DIFFERENTIAL SENSING – CONCEPTS AND APPLICATIONS

The mammalian senses of taste and smell utilize a series of cross-reactive receptors, rather than highly selective receptors. Our group mimics this principle with a series of synthetic and designed receptors for the analysis of complex analytes in real-life settings. The receptors derive from a combination of rational chemical design and modeling, with combinatorial synthesis techniques. Optical signaling often derives either from indicator-displacement assays, or direct modulation of the spectroscopy of the receptor. It will be shown that a union of designed receptors targeted to a class of analytes, with combinatorial methods, gives fingerprints that differentiate between the individual members of the analyte class. The strategy is to use a core-binding element that imparts a bias to each and every member of the library, ensuring affinity of the library members for the class of analytes being targeted. The design of this core derives from standard molecular recognition principles: preorganization, complementary, pair-wise interactions between receptor and analyte, and desolvation. Imparting a bias to the affinity of the library members dramatically reduces the diversity space needed in the library. The fingerprints of the solutions are created using artificial neural networks, principle component analysis, and/or discriminate analysis. The technique represents a marriage of supramolecular chemistry and pattern recognition protocols, and has become known as differential sensing. A variety of examples will be presented, ranging from applications in the biological sciences to commercial beverage analysis.